This is my first tech talk after a long while. It was fun to have an open conversation about the various steps we took to scale our infrastructure and reduce the cost.
The discussion was about the top 10 things we’ve learnt/implemented over the last years.
- Start Small – Starting small when building the infrastructure with AWS.
- Queues solve most of the problems – How queues can solve most of the problems involved in the scaling.
- Dashboards for command control – Building the right dashboards to monitor the big event
- Aggressive scaleup & Careful scaledown – Patterns we used to scale up and down
- Log everything – Logging anything and everything you can
- Reducing Costs – Best practices to reduce the cost while scaling up
- Make it easy to load test – Make it easy to load test the infrastructure easy
- Measure/improve end to end performance – Measuring and building for end to end performance
- Build to fail Circuit Breakers – Building a resilient infrastructure where one failure doesn’t lead to another
- Customer RCA – Building good customer trust even if there are any issues
You can watch the entire video at the yourstory page at https://yourstory.com/session/hackerrank-technology-scale-hiring-aws. You can also download the slides from the talk here at Managing-scale-effectively-at-large-scale-hiring-events-with-AWS.pdf.